This research uses proprietary or administrative data. 

Please contact registrar@uottawa.ca in order to apply for access to that data.

The student-level data used as the basis for this study involves confidential data from individual profiles and was provided to us under an agreement with the University of Ottawa and for the particular purpose of executing this research. For privacy reasons it cannot be published but we are happy to assist those interested in acquiring the data from the institution.

The datasets provided in this dataverse all follow the same structure.

+co_51001.dta
+no2_51001.dta
+o3_51001.dta
+pm25_51001.dta
+so2_51001.dta
+weather_hourly_610578.dta

A row is a daily average of the pollutant noted in the file title, or in the case of weather, each row once again represents a day. 

+ temp_daily_avg 	represents the daily average temperature.
+ precip_daily_avg represents the daily average amount of precipitation (this includes snow).
+ relhum_daily_avg represents the relative humidity.
+ press_daily_avg 	represents atmospheric pressure.
+ winddir_cos_daily_avg is the east-west vector of wind average throughout the day.
+ winddir_sin_daily_avg is the north-south vector of wind average throughout the day. (the intention is to break down a multitude of different ways of breaking down wind into the simplest)
+ date is the year, month, day variable transformed into stata time.

All variables with a "L" prefix represent a lag of that variable, for example L2temp_daily_average represents temp_daily_avg from 2 days prior.

